Fasciation in Lycopersicon. I. Genetic Analysis of Dominance Modification Quentin B

Home , Locule

FASCIATION IN LYCOPERSICON. I. GENETIC ANALYSIS OF DOMINANCE MODIFICATION QUENTIN B. ZIELINSKI' The Blandy Experimental Famz, University of Virginia, Boyce, Virginia Received April 12, 1948 ASCIATIONS are characterized by an increase in weight and volume of F plant tissue and by lack of organized regularity in growth as contrasted with the normal individuals of the same variety or species. As a result, the normal growing point ceases to be a point, but expands into a comb-like structure, often several inches or even feet in width. This definition is primarily based on the morphological classifications of 19th century biologists ( DEVRIES 1894; GORDON1871-72; MASTERS1869; MOLLIARD1897; MOQUIN-TANDON 1841) and on the more recent investigations of WHITE, 1916, 1945; KNOX, 1908; HUS, 1906; and MACARTHUR,1926. The data taken during this investigation lend themselves well to tabulation and have been included for reference. Since marked variability is the rule in the degree of expression of fasciation, such data will permit critical evaluation of the conclusions drawn.

THE PROBLEM AND THE LITEKATURE One of the most striking effects of fasciation in the tomato is the increase in number of locules in the fruit, so that any investigation of fasciation neces- sarily involves the study of locule number inheritance. A preliminary survey of the literature shows that locule number, shape and size are closely associated. Since shape is certainly a character which has been subject to selection in the course of domestication, varieties are therefore likely to be very different in their shape genotypes (minor as well as major). Therefore, shape character- istics should also be recorded as far as possible. In this investigation the genetic symbols have been taken from those of previous workers. The gene symbol for locule number is Lc (few) and IC (many) locules. For shape, o (oval) is used to indicate polar diameter 1.2 or more times the equatorial diameter, 0 (round) for a ratio of 0.95-1.2, and 0' (oblate) for a ratio less than 0.95. The following known gene pairs are involved in this study: D,d standard-dwarf plants (YEAGER) S, s simple-compound flower cluster (YEAGER) 0, o oblate-oval fruits (YEAGER) P,f smooth-fasciated fruit (MACARTHUR) A, a purple-green stem (MACARTHUR) Lj, If leafy-nonleafy inflorescence (MACARTHUR) G, g smooth-grooved fruits (POWERS)

Present address: Department of Horticulture, Oregon State College, Corvallis, Oregon.

GENETICS33: 405 July 1948 406 QUENTIN B, ZIELINSKI In the literature, the papers by YEAGER(1937) and POWERS(1939), give the most significant data on locules and shape inheritance. Linkage data, particularly those of MACARTHUR,show that two genes, located on chromosomes one and five respectively, affect locule number or shape. YEAGER,dealing with chromosome one segregations, concludes that locule number and shape are determined by two separate pairs of genes, but his evidence is not conclusive in this respect. POWERSinfers that he also is dealing with chromosome one segregation, and likewise supports the view that shape and locule numbers are determined by different genes, but the evidence is not conclusive in this case either. Further, it is suggested that he was really dealing in a chromosome five segregation. MACARTHUR(1934) has shown that the recessive gene responsible for fasciation (f) is located on chromosome five which also carries genes for red-green stem (A, a) and leaf inflorescence (Lf,If). POWERS(1939) has published an analysis of locule number inheritance in a cross between L. esculentum strain Johannisfeuer and L. pimpinellifolium (Ked Currant). From his data it would appear that the locule number segregation is determined by a main gene on chromosome one. POWERSfound a strong correlation between oblate fruit and high locule number. A prominent place among early attempts to advance our understanding of the genetics and cytology of fasciation is due WHITE (1913-1916 and 1945). He presents the first analysis of the phenomena stated in the terms of modern genetics. WHITE particularly emphasizes the part environment as well as heredity plays in the production of this character in plants. A comprehensive review of the literature pertaining to the genetics, cytology and morphology of fasciation has been compiled by WHITE (1913, 1916, 1945, and 1948). WARREN’Sdepth-fasciation hypothesis (1924) is of historical interest only. LINDSTROM(1927) disposes of WARREN’Swork by pointing out that WARREN’S hypothesis of complementary action of two pairs of shape factors is not in agreement with the multiple allelic system for the oblate-round-ovate series of fruit shapes. Careful scrutiny of WARREN’Sexperimental data fails to pro- vide convincing proof of his hypothesis. The complementary factorial hypothesis for fruit shape accordingly rests on an insecure basis. MACARTHUR(1944) recognized several rather distinct types of fasciation: 1. Old “Trophy” type. Very flat, or concave with 12-20 locules. 2. “Turban” types. With one or a few central locules and a rim of peripheral locules. 3. Ordinary “fasciated” as in “Ponderosa,” unimproved “Earliana,” and most flat fruited varieties. 4. Fasciated as in “Rouge Nain Hative” (Early Dwarf Red) (with fewer locules). 5. Fasciated stem and extreme fasciation of fruit and stem, stem ribbon-like, obviously composed of three or more stems laterally fused. Fruits often obviously composed of two or three fruits joined into one misshapen concave mass. FASCIATION IN LYCOPERSICON 407 6. The first flower of an inflorescence is often in fasciated varieties the most extremely fasciated and like a “King flower.” 7. Some commercial varieties are dimorphic (Heterozygous Fj Gg) for fasciated (six to 12 locules) and smooth (two to three locules) as in “Tangerine.” MACARTHURthinks that his fasciation gene (j)is present in all the above cases. He has never tried to determine whether there are a series of multiple alleles for the (f) locus or multiple factors modifying it, or the locule gene IC in chromosome one. FRIMMEL(1922) early recognized fasciation in tomatoes. He states that size of fruit is a complex character depending upon number of carpels per fruit and size of individual carpels. In his opinion, fasciation, which is necessary for large fruits, appears to be recessive. HOUGHTALING(1935) in an analysis of size and shape of tomato fruits ob- serves that highly fasciated (multicarpellate) types tend to be both flatter and also larger than bicarpellate types. She found the correlation between size ar d shape to be .3402 _+ .0972 while that between size and number of carpels is .7655+ ,0701. This latter relationship seems to be a mechanical one, for the fasciated types start from a broader meristematic platform.

MATERIALS AND METHODS The material upon which this study is based was obtained from genetic stocks from a number of geneticists as well as from commercial seedsmen. Valuable races of fasciated strains were obtained from H. L. BLOOD,J. W. LESLEY,J. W. MACARTHUR,DAVID WHITE, 0. E. WHITE, and A. F. YEAGER. To all of these contributors the writer wishes to express thanks. The Lyco- persicon genus is especially favorable for genetic analyses due to ease of growing, abundance of progeny, degree of homozygosity and term of growth. Varie- ties and species in this genus are often fertile irtter se, thus facilitating interspecific investigations. During the summer of 1944 125 varieties and strains of L. esculentum were planted, with four replications each, at THE BLANDYEXPERIMENTAL FARM, Boyce, Virginia, and data were recorded for each. All data were submitted to an analysis of variance for significance, and were published by ZIELINSKI (1945). An attempt was made to assemble most of the known many-loculed and likewise the few-loculed types. Crosses were made between 15 fasciated seed parents and ten nonfasciated pollen parents, in all possible combinations, excluding reciprocals. Seed of the f generation were sown in THEBLANDY EXPERIMENTAL FARM greenhouse April 3 and 4, 1945. Approximately 60 plants of each cross were grown in flats and set in a relatively uniform field May 22 and 23, 1945. Only 30 plants were grown of a few crosses due to insufficient hybrid seed.

DESCRIPTION OF FASCIATION IN PATERNAL AND MATERNAL VARIETIES AND SPECIES USED Observations on tomatoes reveal that the main stem axis and all flower parts may be intensely modified. The apex of the stem becomes distorted, flattened 408 QUENTIN B. ZIELINSKI and disorganized (figs. 1, 2, 9). As many as 215 locules have been observed in a single fasciated fruit of the variety Beefsteak. Such monstrous fruits when mature may contain up to 250 seeds, as contrasted with approximately 40 seeds in the normal two-loculed fruits of Red Cherry. The peduncle in these fasciated forms may have as many as 120 vascular bundles, while the normal types such as Ked Currant and Red Cherry possess five. Infinite variation in flower structure abnormality exists in fasciated varieties. The flowers are often split or dialytic. This condition was most prevalent on the variety Triple-L;Crop. Sepals varied in number from a normal of 5 to a maximum 81 (fig. 7).Often the sepals were yellowish colored either at the tip or for a portion of the entire sepal blade (calycanthemy). Various modifications in the number and arrangement of the floral whorls were evident. Calyx and corolla whorls were found existing in one, two or four series with many inter- grades. The whorls frequently appeared to be intimately fused, sepals and petals being recognized only with difficulty. Occasionally sepals or petals were entirely absent. Not infrequently two flowers were found combined and en- closed in a continuous calyx (adhesion). Likewise flowers consisting of only a corolla (or calyx) and a few stamens, growing on the side of and combined with the main corolla (synanthy), were found. Multiplication of whorls (pleiotaxy) involving entire systems occurred. Also polyphylly-or multiplication of number of parts per whorl-was commonly observed (figs. 3, 7). Similarly polyphylly of the androecium was usual. Filaments appeared to be fused to each other (cohesion) and to the walls of corolla and calyx (adhesion). Anther sacs were split into two, three, or four segments or intimately combined. Aborted, shrunken and dwarfed anther sacs were not unusual. In certain instances proliferation had progressed to the point where it was impossible to distinguish between filament and anther sac. Abortion of pollen (contabescence) was found repeatedly. The pistil in some instances was wholly or partially incapable of functioning, owing to various forms of distortion, including gross proliferation, marked staminody, pleiotaxy, and meiophylly of style and ovary (locules). Ovary- locules were so crowded at times, owing to polyphylly, that they were abortive - _-__ __ ~- ______PLATEI Explanation of figures. FIGURE1. MacArthur selection No. 5-strap-shaped apical fasciated stem. FIGURE2. Variety Beefsteak-pronounced grooving and ribbing of stem. FIGURE3. Fasciated fruit of Beefsteak showing twin row of locules. (44 locules). FIGURE4. Variety Brimmer-gross distortion and displacement of locules. (38 locules). FIGURE5. Baldwin I1 (Brazil)-proliferation of floral whorls. FIGURE6. Triple-L-Crop-recurved pedicel accompanied by malformed floral structures. FIGURE7. Cal-1-stem curvature and multiple sepals. FIGURE8. Variety Beefsteak-restricted development of fasciated terminal inflorescence. FIGURE9. Variety Stone-flattened terminal stem with grooving and lateral expansion of stem. 1

hi-

I' FASCIATION IN LYCOPERSICON 409 and rudimentary, resulting in a much distorted fruit (figs. 3, 4). From one to seven pistils (pleiotaxy) were often present in the same flower, sometimes all capable of functioning; in other cases all but one, or sometimes all, abortive. Sterility was present and such specimens failed at anthesis. In the majority of cases, however, examination of a mature fruit demonstrated fertility to be almost perfect. The gynoeciums in general seem to be the most altered and the calyx the least so. The stems, peduncle and pedicel were notably distorted. However, this condition is achieved through a gradual progression from a normal seedling to the fasciated more mature plant. The young plants possess normal phyllotaxy, contour and lineation. It is not until the plants reach about six to eight inches in height (for most varieties) that disarrangement begins to be noticeable (fig. 9). The stem width increases in extent and the leaves in number as the plant approaches sexual maturity. Extremely fasciated varieties usually first set a highly fasciated terminal flower (figs. 7, 8). Occasionally such flowers set fruit and develop monstrous specimens (figs. 3, 4). More frequently, the terminal flower aborts and eventually abscisses or becomes desiccated. Occasionally terminal flowers apparently become fasciated earlier in their ontogeny than usual and even fail to differentiate into a recognizable floral pattern. This is interesting for it illustrates one of the extreme modifications of floral patterns achieved through intense precocious fasciation. This situation, furthermore, is peculiar in that this highly fasciated terminal first flower is borne upon a broadly flattened fasciated pedicel and peduncle. The flower is characteristically arched, forming a complete recurvature to the supporting pedicel (fig. 6). The peduncle frequently bears numerous other flowers, located on a very irregular phyllotaxic plan, that are apparently normal. These possess plump anther sacs filled with abundant very immature pollen grains (meiosis is successful) and apparently normal pistils. Often ten to 14 such flowers will be found having normal pedicles on a single fasciated peduncle. Such flowers, however, do not develop further due, presumably, to vascular difficulties involved in the fasciated peduncle. .- The stem of a mature plant has a cylindrical base, which gradually widens, developing the characteristic flat, ribbon-like, fasciated condition. When fasciations in the stem are pronounced or extreme, a distinct grooving and ribbing of vascular strands is clearly evident. Stems vary, for example, from a normal width of 5/16ths of an inch to 7/8ths of an inch. Generally, such lateral expansion is accompanied by a corresponding decrease in stem thickness although the extent of this is only partially compensatory. The stem is generally straight, occasionally being slightly curved due to irregularities in growth (figs. 7, 9). It is characteristically unbranched even when a length of as much as eight inches is involved. This curious behavior adds to the grotesqueness of the plant in the field. The fasciated tomato fruits are highly variable, even among the members resulting from a given inflorescence. There is a striking tendency for fasciated fruits to be associated with oblateness, that is, markedly flattened or depressed 410 QUENTIN B. ZIELINSKI at the poles. However, the variety Oxheart is a conspicuous exception, there being seemingly a somewhat looser relation here between locule number and fruit shape. The author has noted many of the types described by MACARTHUR (1944). Conspicuous among fasciated patterns is that of the “Trophy” type, wherein the fruit becomes very flat, markedly compressed at the stem attach- ment point, and may possess from ten to twenty locules. The so-called “Tur- ban” type is characteristic for the varieties White Beauty, Triple-L-Crop, and Baldwin 11. A commercial variety of lesser importance, possessing such char- acteristics, has the name “Turk’s Turban.” Such fruits have three or four central protruding locules with a rim of peripheral locules-representing in

TABLE1 Average number of jloral parts per whorl above normal number for genus, for 100 flowers of each tomato variety, July, 1944

~ __ __ ~ ~ ~ __ ~ __ SEPALS PETALS STAMENS LOCULES ______~ ___. Normal 5.0 5.0 5.0 2.0

Red Currant 0.0 0.0 0.0 0.0 Italian Canner 1 .0 0.9 1 .0 0.0 Red Plum 0.5 0.6 0.4 0.1 Red Peach 1.4 1.5 1.3 0.1 Sugar 0.4 0 . .i 0.4 0.2 Yellow Peach 0.8 1 .0 1 .0 0.2 Yellow Pear 0.7 0.9 0.5 0.2 Baldwin I 0.2 0.2 0.2 0.2 Break 0’ Day 1.8 2.0 2.6 6.5 Oxheart 3.5 3.9 5 .9 7.5 Triple-L-Crop 3.8 2.0 4.5 10.2 Pritchard 1.2 1.3 1.7 4.2 Brimmer 7.9 6.1 5.6 14.4 Stokesdale 1.3 1.6 1.7 6.7 Matchless 1.8 2.3 3.0 6.5 Stone 4.8 5.6 6.9 11.4 C o1 ossa 1 8.8 9.7 11.4 11.6 Oxheart 2.1 2.1 3.8 8.7 Bison 2.9 2.9 2.6 6.9 Ponderosa 3.6 3.7 3.8 8.7 Orange King 2.0 1.9 2.3 3.4 Truckers’ Delight 0.2 0.4 0.3 1 .x Earliest and Best 2.1 2.1 2.2 3.7 Rutgers 0.6 1.1 1.7 3.6 Beefsteak 4.4 4.6 5.2 9.4 Prodigious 5.7 6.5 8.9 8.1 White Beauty 5.2 5.8 6.1 8.1 Marglobe 1.5 1.8 2.1 2.7 Summerset 1.8 2.2 2.1 2.4 Golden Queen 1.4 1.6 2.0 3.5

Average 2.43 2.62 3.05 5.17 -. FASCIATION IN LYCOPERSICON 41 1 the tomato a shape comparable to the turban type squash. Another fasciated type represented in this study is the “Beefsteak” type, found in varieties such as Ponderosa, Brimmer, Beefsteak, Santa Clara Canner, Cal-1, and Cal-2. The fruits are large to extremely large. They are usually rather flat but fairly deep and oft& oval in transverse cross section. They are sometimes smooth but more often corrugated with a deep stem-end scar and a pronounced blossom- end scar and depression. Frequently the fruits are lobed and doubled. The outer walls and partition walls are very thick. The locules are numerous (12- 20) and irregular and are well filled with pulp, containing many to few seeds. The central mass is very large, firm and fleshy. Occasionally this type prolifer- ates extensively yielding fruits of monstrous appearance (figs. 3, 4). While taking notes on arrangements of floral structure the question of ac- curately scoring fasciations arose. In order to determine the consistency of relationships among the floral whorls data were collected on individual whorls for comparison. The results for thirty varieties are shown in table 1. By inspec- tion of table 1 the average number of parts per floral whorl in the fasciated types is seen to be a progressive increase as one advances from the outer toward the innermost floral whorl. The data are given as the average number of floral parts per whorl above that considered normal (5-5-5-2) by the author for the genus. Hence, in order to represent actual conditions the sepal numbers should be increased by five, petal numbers by five, stamen numbers by five, and locule numbers by two. The data as listed, however, are on a comparable basis and the calyx is thus seen to be the least, and the gynoecium the most, TABLE2 Aalerage locde number of 36 tomato varieties grown at The Blandy Experimental Farm, July, 1944. - I____~_..__ ~ ____ LOCULE LOCULE VARIETY VARIETY NUMBER NUMBER --_-___ 1. Red Currant 2.03 19. Pritchard 5.74 2. Italian Canner 2.03 20. Matchless 7.66 3. Red Pear 2.05 21. Break 0’ Day 7.68 4. Yellow Cherry 2.06“- 22. Bison 8.98 5. Red Plum 2.10 23. Prodigious 10.05 6. Sugar 2.12 24. White Beauty 10.18 7. Yellow Pear 2.21 25. Oxheart 10.76 8. Baldwin I 2.27 26. Ponderosa 11.40 9. Ked Peach 2.49 27. Beefsteak 11.74 10. Yellow Peach 2.50 28. Triple-L-Crop 12.42 11. Truckers’ Delight 3.85 29. Brimmer 12.99 12. Summerset 4.32 30. Colossal 13.61 13. Stokesdale 4.53 31. Stone 13.74 14. Marglobe 4.81 32. Baldwin I1 14.58 15. Orange King 5.41 33. Cal-2 15.33 16. Golden Queen 5.58 34. Santa Clara Canner 15.90 17. Rutgers 5.61 35. Trophy 15.98 18. Earliest and Best 5.67 36. Cal-1 16.05 412 QUENTIN B. ZIELINSKI affected. There is a progressive increase in expression of abnormality as one examines first the sepals, then petals, then stamens and finally the locules. As WHITE (1916) pointed out, this progression in the manifestation of abnormal condition is in accordance with the observations on other parts of the plant, for example stem structure. During the summer of 1944 seeds of 125 varieties and selections were sown in the greenhouse. Thirty plants of each were grown and transplanted to a relatively uniform field during the second week in May. Observations during this season were aimed at (a) establishing the extent of fasciation occurring in commercial varieties by measuring locule number of transversely halved fruits, (b) to observe whether or not segregation was occurring among individuals of a population of a commercial variety, and (c) to measure the effect of seasonal influence on the degree of fasciation. Jn addition, pollinations involving 6842 flowers were made. Observations were made during each of the months of July, August and September. This period was especially dry, and thus the fullest expression of the fasciation genes may not have been obtained. However, en- vironmental conditions were similar for all varieties involved. The varieties studied, the majority of which were commercial varieties, exhibited a wide range in mean locule number, from two to as many as 14 or 16 (table 2).

Summarized frequency distributions of locule numbers are arranged in tables 3 and 4. The mean monthly values cited are based on observations on ten flowers on each of ten plants of each variety. There was a striking trend toward lower locule number as the season advanced. This was exhibited in every one of the varieties grown; even in the case of the three varieties (Oxheart, Brim- mer and Stone) in which readings for August were slightly higher than those for July. There was a very marked decrease in locule number in feptember. Analysis of variance indicated that the difference in mean locule number between months within a variety, as were those between varieties, was highly significant. The end-of-season effect was most pronounced in the extremely fasciated varieties; this was clearly brought out by a highly significant interaction between varieties and months. A preliminary examination of the data gave no evidence of segregation for locule number between plants within a variety. Analyses of variance were calculated for three of the more highly fasciated varieties, in which segregation might be expected to .be more easily detected than in low locule-number varieties. There were no significant differences between plants within these three high locule-number varieties. Within the limitations imposed by the restriction of observations to ten plants in each variety, it may therefore be concluded, that each of the horticultural varieties included in this study is characterized by a fairly uniform genotype in respect to the fasciation character. In striking contrast, however, are the very significant differences between months. It is evident that in the study of the inheritance of fasciation the influence of the environment is very important. In no cases were plants found producing onZy normal two-loculed fruits. FASCIATION IN LYCOPERSICON 413

33 CIN 414 QUENTIN B. ZIELINSKI It I

v;N

h .gP 5

Tm w F9 H r3 2 3 4

N CI .-)

0 3 m

N FASCIATION IN LYCOPERSICON 41

4 3

-3

3 "N m - 3 416 QUENTIN B. ZIELINSKI

NWOC v, -COT.. -3m 9 3.3 2

i 3

3 ,- 3

e N

3 "; d.

v: 00

,-N LO a h em 3P LO r- ,.s- m vU mI w r-"3 U mlnm m +3 3

m lr, - ch * m -*

3 FASCIATION IN LYCOPERSICON 417

w2: U E 2 @.5: -a81 2: .-s a .% Pb va .c 5% 2.9 2:: ;$-R :cl %S22 3 f- $2 x uu CI .C22 ; 6% >am ;a"* $$ h .$ 23g- -3%

'Ns U E

...... 418 QUENTIN B. ZIELINSKI The hybridization program included the maximum possible number (150) of intervarietal and interspecific crosses between the 15 fasciated and ten nonfasciated two-loculed types, using the nonfasciated varieties as pollen parents and medium to high fasciated varieties as maternal or seed parents. This arrangement was followed for convenience. Abundant pollen was always assured in the anther sacs of nonfasciated varieties. This was not true for the fasciated varieties, but the latter generally possessed a much greater stigmatic area for pollenizing and emasculation was much less a problem. Of the 6842 pollinations, 621 were successful, a total of 9 percent. The pollen parent giving the highest percent of sets among the 150 combinations was Red Pear, setting 12 percent of the pollinations. Among the seed parents Santa Clara Canner ranked highest, 12.4 percent, in fruit set. The cross Santa Clara CannerXSugar gave the highest set (13 percent). What significance if any may be attached to these observations is not now very evident.

DESCRIPTION OF F1 GENERATION Observations during the 1945 season were aimed principally at (1) measuring fruit locule number and (2) fruit size. In addition, ten flowers were selfed for each F1 family. Data were recorded between August 15 and September 1. Growing conditions during the summer of 1945 were especially favorable for the maximum expression of the fasciation character. Rainfall was abundant and warm growing weather prevailed. Locule numbers were determined by transversely cutting fruits of various degrees of maturity ranging from those shortly after blossom fall to fully ripened fruits. It was noted that with a careiul examination a high degree of accuracy could be obtained even from very immature fruits. Data were taken on a population of 150 fruits selected at random from the 60 (30 in a few cases) hybrid plants of each cross. These were sectioned in the field and notes made immediately. Fruit size and shape were determined by methods of YEAGER(1937) and LESLEY(1927). Polar and equatorial diameter measurements were made with calipers expressed as a ratio for parental and hybrid populations. Equatorial diameter = Shape Index. Polar diameter Summarized data for locule numbers are given in table 5. This table is arranged to indicate values for pollen and seed parents, and their F1 averages. F1 ratios and F1 spread or range data are given in table 7. Parenta1:Fx ratios were calculated by comparing the two parental values with that of their hybrid,* as shown in the following example:

Marglobe 0 ...... 4.80 locules Red Currant 8...... 2.03 locules Difference...... 2.77 locules .47 t2.77= 17% Ratio of FIto 0 and parents FI...... 2.50 locules Red Currant...... 2.03 locules Difference...... 47 locule FASCIATION IN LYCOPERSICON 419

REDCURRANT F; MAFGLOBE 2.03 2.5 4.6

17% OF SPREAD BETMEN PARENT

This ratio (.17) may be interpreted as an effect of superimposing Red Currant upon Marglobe, resulting in a shift of the F1 locule number of 17 percent from that of the pollen parent toward that of the seed parent. These pooled pollen parent F1 ratios are graphed for convenience in table 7, while conversely, pooled seed parent F1 ratios are given in table 8. Data showing the spread or range of the F1 distribution are given in table 6. These are calculated as the difference between the high and low locule numbered fruits from 150 specimens of each cross. TABLE5 Summary of locule number data for parental and F1 tomato progenies, indicating parental averages and Fl averages fw 150 fruits oj each cross. _____ ._- RED ITALIAN YEL- YEL- POLLEN RED RED BALD- RED CUR- CAN- LOW SUGAR LOW PARENTS PEAR PLUM WIN I PEACX RANT NER CHERRY PEAR PEACH _._____ Seed Paroils 2.03 2.03 2.05 2.06 2.10 2.21 2.26 2.27 2.69 2.99 AVE.

Marglobe 4.8 2.5 3.1 3.2 3.2 3.6 2.5 3.3 3.6 5.1 3.8 3.39 fritchard 5.7 2.6 3.4 3.6 3.3 3.8 2.7 3.6 3.4 4.1 4.8 3.59 Bison 9.0 2.5 3.4 3.3 3.2 4.3 2.4 3.3 3.5 3.6 4.3 3.38 White Beauty 10.18 3.6 3.1 3.6 3.2 3.3 3.2 3.5 3.2 4.8 4.9 3.7 Oxheart . 10.8 3.0 3.7 3.8 3.3 4.2 3.1 3.1 3.9 3.6 4.3 3.66 Ponderosa 11.4 2.7 3.9 4.1 3.3 4.3 3.1 3.8 3.4 4.7 5.2 3.84 Beefsteak 11.8 2 I 3.9 3.1 3.4 4.7 3.1 3.3 3.6 4.2 5.2 3.8 Triple-L-C. 12.4 2.5 3.7 3.1 3.5 3.5 2.8 2.6 3.6 3.7 4.1 3.4 Rrimmer 13.0 2.9 3.9 4.1 3.8 4.8 3.2 3.3 4.3 3.2 5.2 3.8 Colossal 13.6 2.5 3.9 3.5 3.4 4.1 3.2 3.6 3.7 3.8 4.0 3.5 Baldain I1 14.6 2.6 3.4 3.4 3.4 3.6 3.1 3.5 3.7 3.8 3.5 3.4 Cal-2 15.3 2.8 3.8 3.9 4.1 4.5 3.4 3.5 3.8 4.9 5.2 4.0 Santa Clara C. 15.9 3.0 3.9 3.8 3.3 4.4 3.3 3.5 3.1 4.6 4.5 3.8 Trophy 16.0 2.8 3.4 3.5 3.3 4.2 3.3 3.6 4.0 4.6 5.0 3.1 Cal-1 16.0 2.9 3.8 4.2 3.5 4.9 4.4 3.9 3.8 4.0 3.8 3.9

F I Averages 2.7 3.7 3.1 3.4 4.2 3.1 3.5 3.1 4.2 4.5 I

The author wishes to acknowledge the contribution of DR. D. B. DELURY,Associate Statis- tician, VIRGINIAPOLYTECHNIC INSTITUTE, in suggesting this expression of results.

Examination of data in table 5 indicates that non-fasciation in tomatoes is partially dominant. The average F1 locule number always exceeded the nonfasciated parent but never exceeded this parent in locule number by more than 2.80 locules on the average (Red Plum (2.10)XCal-1 (16) (F14.9). The fasciated parents always exceeded their F1 values except in the cross Marglobe XRed Peach. Table 5 also indicates that the closest F1 locule number association was between Marglobe and its F1 hybrids (3.39). Here this fasciated variety (4.8) exceeded the F1 average by 1.41 locules. The widest locule number 420 QUENTIN B. ZIELINSKI TABLE6 Siu"ary of locule laumber data for purentul and F, tomato progenies, indicatine Parental-F, ratios and FI spread for 150 fruits q' each cross.

.___- ._ ... ~~

RED ITALIAN RED YELLOW RED SELLOW BALD- RED YELLO\\ SUGAR URRANT CANNER rEhR CHERRY PLUM PEAR WIN I PEACH PEACH

Marglohe Ratio ,I7 .38 .04 .41 .5s .t1 .40 .52 .14 .45 .42 Spread .0 6.0 2.0 1.0 4.0 2.0 3.0 2.0 5.0 3.0 3.2

Pritchard .15 .37 .O1 .33 .47 .11 .39 .33 .6i .66 .35 2.0 3.0 3.0 4.0 .i. 0 2.0 4.0 4.0 5.0 5.0 3.7

Bison .07 .19 .1R .16 .32 .03 .30 .18 .11 .22 .18 3.0 3.0 .3.0 3.0 5.0 3.0 3.0 3.0 3.0 5.0 3.4

White Beauty .I9 .20 .31 .14 .I3 .I2 .16 .12 .28 .27 .19 2.0 3.0 40 2.0 4 . 0 3.0 2.0 3.0 5.0 5.0 3.3

Oxheart .I1 .19 .20 .14 .24 .I1 .t7 ,19 .ll .17 .16 3.0 2.0 4.0 2.0 5.0 1 .0 3.0 4.0 5.0 3.0 3.2

Ponderosa .Oi .20 .22 .13 .24 .lo .16 .12 .23 .26 .17 2.0 2.0 5.0 4.0 5.0 3.0 4.0 3.0 5.0 i.O 4.0

Beefsteak .07 .t9 .17 .38 .2i .09 , 17 .14 . li .25 .19 2.0 4.0 3.0 3.0 i.0 3.0 3.0 4.0 5.0 7.0 4.1

Triple-L-Crop .os .t6 .15 .t4 .14 .06 .10 .15 .1t .12 .12 2.0 6.0 3.0 4.0 3.0 3.0 4.0 3.0 3.0 6.0 3.7

Brimmer .ox .17 .18 .16 .24 .09 .03 .19 .05 .22 .14 2.0 6.0 3.0 3.0 6.0 4.0 3.0 5.0 5.0 7.0 4.4

Coloqsal .04 .16 .12 .12 . 17 .09 .09 .13 .10 .10 .it 2.0 3.0 5.0 3.0 4.0 3.0 4.0 2.0 4.0 5.0 3.5

Baldwin I1 .os .11 .I1 .t1 .12 .07 .tl .t8 .09 .04 .w 1.0 3.0 3.0 4.0 5.0 3.0 3.0 4.0 3.0 3.0 3.2

Cal-2 .06 .13 .13 .l5 .18 . 09 .09 .12 .li . 17 . 1 3 2.0 3.0 4.0 6.0 4.0 3.0 3.0 3.0 6.0 7.0 4.1

Santa Clara .06 .13 .12 .09 .1i .08 .09 .11 .1& .17 .ll Canner 2.0 4.0 4.0 2.0 5.0 2.0 2.0 4.0 4.0 6.0 3.5

Trophy .06 .09 .49 . 00 .15 .ox .10 .13 .12 .15 .14 2.0 4.0 2.0 4.0 4.0 3.0 3.0 3.0 4.0 5.0 3.4

Cal-1 .06 .13 . t5 .10 .20 .16 .12 .11 .10 .06 .19 2.0 3.0 5.0 3.0 5.0 4.0 5.0 4.0 3.0 5.0 3.9 __ AV. Parental-F1 Ratio .OX5 ,186 .l7 ,116 .21 .09 .16 .18 .29 .22 AV. F, Spread 2.0 3.6 3.5 3.4 4.7 2.8 3.3 3.4 4.3 5.3 association occurred between Cal-1 (fasciated) X all nonfasciated varieties with a difference of 12.1 locules on the average. Ked Currant (L.pimpinellifolium) and Sugar (L.esculeiztum) show the highest degree of dominance among those selected as nonfasciated parents in this study (table 7). The varieties Red Peach and Yellow Peach, both possessing the highest locule number among the nonfasciated parents, also exhibit the least dominance. Between these two extremes no trend is observable, but it FASCIATION IN LYCOPERSICON 42 1 TABLE7 Percent dominance modification of locule number for FI of nonfasciated (pollen parent) by fasciated (seed parent). .~ -~______._ ~ ~~_____~ _____

NON-FASCIATED POLLEN PARENTS %

1. Red Currant 8 2. Italian Canner 18 3. Red Pear 16 4. Yellow Cherry 17 5. Red Plum 25 6. Sugar 9 7. Yellow Pear 16 8. Baldwin I 16 9. Red Peach 18 10. Yellow Peach 16

~~ ~~ -

may be noted that three rather definite groupings are to be seen. Two types, Red Currant and Sugar, gave less than ten percent dominance while Ked Plum exhibited about 25 percent and the remaining varieties fall between 15 and 20 percent. Inability to make further deductions may possibly be due to inherent experimental error or complex interaction. In table 8, it is evident that among the fasciated varieties most modified by the dominance of nonfasciation, Marglobe and Pritchard were strikingly affected, being 42 percent and 35 percent respectively. It is to be noted, too, that these varieties have the lowest locule number among the fasciated varieties. Least dominance modification (.09 percent) is shown by Baldwin I1 (14.6 locules). All highly fasciated varieties (Triple-L-Crop, Brimmer, Colossal,

TABLE8 Percent dominance modijication of loctde Number of 171 of fasciated (seed parents) by non-fasciated (pollen parents).

~ ~~____ - _____~ ___-~ - - ~~~ FASCIATED SEED PARENTS %

1. Marglobe 42 2. Pritchard 35 3. Bison 18 4. White Beauty 19 5. Oxheart 16 6. Pon+rosa 17 7. Beefsteak 19 8. Triple-L-Crop 13 9. Brimmer 14 10. Colossal 12 11. Baldwin I1 9 12. Cal-2 13 13. Santa Clara Canner 11 14. Trophy 15 15. Cal-1 12

~~ 422 QUENTIN B. ZIELINSKI Baldwin 11, Cal-2, Santa Clara Canner, Trophy and Cal-1) were modified only slightly, indicating perhaps that in addition to one or two major genes, the highly fasciated varieties possess numerous modifying genes that determine the degree to which the character is expressed in fruits, flowers, and stems or on which polygenic systems are operative. A comparison of fruit shape for parental and F1 progenies of the fasciated and nonfasciated varieties involved in this study are given in table 9. The parents have been arranged in order of increasing magnitude from the most

TABLE9 Comparisoit of fruit shape for parental and FIprogenies involving fasciated and non-fasciated varieties. equatorial diameter Ratio of __---___ --shape index polar diameter

PALIAN YEL- YEL- RED YEL- RED EED BALD- RED CAN- LOU LOW CUR- SUGAR mw PEAR PLUII WIN I PEACH NER PEAR CHERRY RANT PE VH

.40 .61 .63 .64 .85 .89 1.01 1.03 1.21 1.22

Oxheart .80 .90 .94 .90 .82 ,02 .90 ,85 .95 .82 .89 .89 Marglohe 1.40 -.94 1.24 1.26 1.36 1.19 1.30 .96 1.10 1.22 1.33 1.19 Pritchard 1.41 1.33 1.20 1.26 1.58 1.30 1.33 1.06 1.24 1.58 1.55 1.34 Ponderosa 1.87 1.31 1.18 1.20 1.48 1.26 1.45 1.46 1.15 1.48 1.53 1.35 Beefsteak 1.91 .82 1.38 1.29 1.44 1.3R 1.26 1.20 1.06 1.44 1.33 1.26 Brimmer 1.92 1.03 1.14 1.1) 1.48 1.13 1.33 1.06 1.17 1.48 1.50 1.25 W. Beauty 1.98 1.29 1.23 1.35 1.03 1.20 1.07 1.00 1.07 1.00 1.0i 1.30 Triple-L.-C. 2.01 1.22 1.20 1.03 1.14 1.29 1.37 1.08 1.18 1.14 1.47 1.21 Santa C.C. 2.23 1.52 1.30 1.20 1.28 1.20 1.22 1.20 1.25 1.28 1.60 1.31 Colossal 2.26 1.04 1.07 1.07 1.28 1.00 1.60 1.06 1.37 1.28 1.55 1.23 Cal-2 2.43 1.23 1.39 1.00 1.25 1.38 1.42 1.13 1.16 1.25 1.61 1.28 Cal-l 2.44 1.10 1.10 1.07 1.34 1.28 1.18 .74 1.30 1.34 1.89 1.23 Baldwin I1 2.50 1.27 1.08 1.00 1.20 1.20 1.00 1.00 1.17 1.20 1.76 1.19 Trophy 2.51 1.19 1.22 1.07 1.46 1.25 1.24 1.08 1.22 1.46 1.41 1.26

Average - - ovate to the most oblate shape. Examination of the F1 values gives no indication of trends. In most instances the F1 was more ovate than the fasciated parents and more oblate than the nonfasciated parents. A few exceptions to this generalization will be noted. Oxheart was notable for fruit shape among its progeny; strikingly uniform and always slightly ovate. Rather unusual too was the finding that F1 progenies involving Ponderosa were the most oblate (1.35), yet Ponderosa is not more strikingly oblate than were some of the other parents. The most oblate nonfasciated variety, Yellow Peach (1.22), gave as expected the most oblate offspring (average of 1.50). The most ovate nonfasciated parent, Italian Canner (.40), yielded rather oblate offspring (1.16) in contrast to Red Currant (1.01) which gave F1 progeny measuring 1.07. It might be postulated that such fruit-shape types as Oxheart, Pear and Plum may be determined by genes at several loci and there may be distinct dominant fruit lengthening genes, that is, genes which make the fruit more ovate. FASCIATION IN LYCOPERSICON 423

DISCUSSION In order that the pertinent literature bearing on the problem of inheritance of fasciation, when analyzed in the light of the modern interpretation of the phenomena, can be correlated a review of the genetic evidence will be presented. Fasciations in plants exhibit a series of phenomena, the more striking being increase in weight and volume and disorganization of regular growth patterns. As a result, growing points, whether they be roots, stems, flowers or fruits, expand into structures more complex than those characteristic for the taxo- nomic category. In Lycopersicon the main stem axis and all floral parts may be intensely modified. Cultivated tomatoes exhibit a wide range in locule number, from two to 215 having been observed. Those forms with a higher locule number than that orthodox for the family are described as fasciated, especially so if the high number is associated with irregularity of outline of the fruit. There can be no doubt that at least two pairs of major genes, located on chromosomes 1 and 5 respectively, affecLlocule number and shape, and it is important to distinguish between them because they may not represent identical duplicates, but may differ markedly in potency. In fact, the records in the literature do support the supposition that only the gene on chromosome 5 (MACARTHUR)is a true “fasciation” gene (implying multiplication of parts in entire floral systems) and it seems possible that this gene is recognizable by its effect in making the fruit contour irregular (so-called) rough versus smooth or regular. On the other hand, the gene on chromosome 1 appears to increase locule number from the basic two to three only as far as about five to eight, and does not markedly disturb fruit outline regularity. Yet it cannot be dismissed from consideration if locule number is ‘to be the primary subject of scoring, especially in view of the possibility that this gene may, under certain genotypic conditions, be more extreme in its expression. Nevertheless, it is suggested that it will be a good working hypothesis to assume that the two major gene pairs can be distinguished by their effect on fruit irregularity. Chromosome 1 Genes YEACER(1937) presents evidence that locule number is associated with weight and shape (breadth!length). There is a closer association between locule number and shape than with weight, and examination of partial correlations shows that shape and weight are really only associated because of their mutual association in locule number. YEAGER’Sdata, however, seem to indicate that the association between Lc and 0 is developmental (pleiotropic- that is, action of one gene on two characters simultaneously) and not caused by genetic linkage, since high locule number is associated with oblateness even where the parental association is the reverse of this (Progeny A in table 1 and Progeny G in table 3). YEAGERhas overlooked the fact that, on his linkage interpretation, both of these crosses should have shown a negative rather than positive correlation between locule number and shape (coupling instead of repulsion). 424 QUENTIN B. ZIELINSKI YEAGERsignificantly points out that strain KO. 15, although many-loculed, is round and pointed instead of oblate; and when crossed with oblate, few gave some ovals in Fz. From this he concludes that it really carries oval, that is, is a coupling combination instead of a repulsion combination, and that the shape effect of the oval gene is offset by the flattening effect of high locule number. But the phenotype of Strain No. 15 and the other results which he cites are better explained on the basis of an independent shape genotype, no doubt multifactorial, and superimposed on the main locule number-shape gene. YEAGER’Sdata indicate that in some crosses at least, low locule number (Lc)is highly dominant. His usage of the 0 symbols (shape), o-oval, 0-round, and 0’ oblate, implies, on the other hand, that the flatter types (oblate or round) with which the plurilocular condition is associated, are dominant over the longer (oval) type. Unfortunately, he does not give any data from which conclusions can be drawn. Data presented in table 5 lend support to YEAGER’S conclusion that low locule number (Lc) is highly dominant. However, it is shown that this is not invariably true. Data from this study indicate no cases of complete dominance for non-fasciation or low locule number. The average F1 locule number always exceeded the value for the non-fasciated parent but never by more than 2.80 locules on the average. LINDSTROM(1937) presents some interesting data in this connection. The Fz’s as depicted in his fig. 1, while showing that dominance is usually in a direction of oblateness, show also that dominance is susceptible to much modification. The following data from his table 10 are of considerable interest in showing almost complete reversal of prepotency.

OBLATE ROUND Fi PARENT PARENT

Bonny BestXYellow Cherry 1:38 1 :09 1:11 Bonny BestXYellow Peach 1:32 1:09 1:26

Data presented in table 9 of the present paper while referring to F1 progeny, strikingly demonstrate the general association of shape and locule number or fasciation. In most instances the FI was more ovate than the fasciated parents and more oblate than the non-fasciated parents. Of interest, however, is the indication that the direction of dominance is not always in the direction of oblateness as in the case of Oxheart, Pear and Plum shapes. Shape in these varieties may be determined by genes at several loci and there may be distinct dominant fruit lengthening genes, which make the fruit more ovate. If various strains of Lycopersicum esculentum differ in their minor shape genotypes, there is every likelihood that their minor locule number genotypes are also different, though the difference in this respect may not be so great since locule number is much less subject to conscious selection than\fruit shape. FASCIATION IN LYCOPERSICON 425 LINDSTROM(1928) drew attention to the high positive correlation between the flattened shape and large size. His data were all derived from crosses in which the more oblate parent was of larger size, and he, therefore, hesitated to decide between the alternative of linkage or pleiotropy (morphological association) until he had studied crosses in the reverse association. YEAGER(1937) has reported on such crosses in which the more oblate parent was of smaller size (and sometimes with more numerous locules), and in these there was still a positive correlation, indicating that the pleiotropic interpretation is to be preferred. YEAGER’Sfurther analysis of partial correlations is particularly in- structive. He found that in some families which gave significant positive correlations between weight and shape, significant negative correlations occurred between these attributes when locule number was eliminated. YEAGERinter- prets this as throwing considerable doubt on the validity of the assumption that oblate shape determines large size. He concludes that weight and shape are only associated because of their mutual association with locule number, and brings forward a hypothesis involving the association of oblateness and small lqcule number. A much more plausible interpretation of his reversal of the usual weight-shape correlation on elimination of locule number is that large size genes other than those which affect size by way of locule number are fundamental fruit lengthening genes-that is, genes which make the fruit more ovate. We thus arrive at the conclusion that fruit shape is determined by two sets of mutually antagonistic size factors-(a) general size genes, which primarily affect longitudinal growth, and by (b) chromosome 1 genes which partially determine increase in locule number and within limits the amount of lateral “filling OU~”of the fruit. It is likely that there are, in addition, other general size genes which have no effect on shape at all.

Chromosome 5 Genes It is evident from his description that the character which MACARTHUR (1926) describes as fasciation in the Albino Ponderosa strain is true fasciation. It has a “distinctly lobulated irregular contour, the irregularity extending to the extreme of producing mmpound fruits by fusion of as many as four. This roughness of outline is anticipated even in the flower by a strongly correlated flattening of the pistil, a high number of petals and sepals (eight to ten or more instead of five to seven) and by a very large number of ovarian locules.” Later MACARTHUR(1934) showed that the recessive gene responsible for fasciation (f) is located on chromosome 5 which also carries genes for red- green (A, a) and leafy inflorescence (Lf,y). From MACARTHUR’Sstatement,f (fasciation) and o (elongated fruit) show a complex interaction which defies analysis in progenies where both are segregating. POWERS(1939) has published an analysis of locule number inheritance in a cross between L. esculentum strain Johannisfeuer and L. pimpinellifolium (Red Currant). By inference it is made to appear that the locule number segregation is determined by a main gene on chromosome 1, since he locates a locule 426 QUENTIN H. ZIELINSKI number gene of considerable potency midway between the shape gene 0 (short fruit) (which is known to be located on chromosome 1) and a gene pair G, g controlling the difference between smooth and grooved fruit. This gene has not been described in detail by POWERSor any other worker, and the presumed linkage (of about 15-18 percent) is only demonstrated in the coupling and not in the repulsion phase. In such a wide interspecific cross of this type, the avail- able data can be interpreted equally well as indicative of a pleiotropic association of grooving with shape and locule number. The apparent crossover between grooving and shape here is the result of minor gene segregation. Not until it has been demonstrated in the repulsion phase can such an association as this be accepted as genetic linkage. So far as his shape segregation (0, 0) is concerned, there is no genetic evidence that the segregation refers to chromosome one. The Johannisfeuer parent with grooved fruits has a locule number of 9.6, and this, together with a photograph of a transverse section of fruit which shows a flattened placenta, is strongly indicative of fasciation as interpreted in this study. POWERSfound a strong correlation between oblate fruit and high locule number. Since the gene on chromosome 1 which affects locule number also affects shape apparently by reason of morphological associations, it seems highly probable that the shape segregation which POWERSscored was an expression of the fasciation or locule number gene on chromosome 5. In connection with this hypothesis that locule number, shape and grooving are pleiotropic effects of the same gene, it issignificant that the two apparent crossover classes have identical locule numbers. POWERSinterpreted this as indicating that a locule number gene was situated midway between the shape and grooving genes. My interpretation is that these two groups represent intermediates between the F (non-fasciation) and f (fasciation) genotypes whose distinction is obscured by the concurrent segregation of other minor genes, and any separation of them into two approximately equal sized groups is likely to be purely at random insofar as the main genotype is concerned. POWERS’F1 and Fz data demonstrated that shape shows little if any dominance, while the few loculed character is highly dominant over the many loculed (fasciated) one. Evidently, the dominance situation in this interspecific cross is nearly the same as in some crosses within L. esculerttum. There is evidence that the shape genotype is highly variable, more so between varieties within 1;. esculentum than between the two species, and there is, therefore, every likelihood that the genetic basis for fasciation, at least in some of it aspects, differs from strain to strain.

SUMMARY Fasciation in tomatoes affects primarily the floral whorls and the terminal stem. The apex of the stem becomes distorted, flattened, disorganized and recurved upon itself. High loculed fruits occur at infrequent intervals. While the variety Beefsteak generally possesses ten to 12 locules, many specimens have been noted which have 60 to 100 locules and one specimen with 215 locules FASCIATION IN LYCOPERSICON 427 was observed. Such monstrous fruits may contain a very high seed number, up to 250 seeds in one instance. All floral whorls are affected by fasciation. Abortion of pollen was found repeatedly. The distorted pistils in many instances were wholly or partially incapable of functioning. From one to seven pistils, both functional and abortive, were often present in the same flower. Sterility was present and such specimens failed’at anthesis. In the majority of cases fertility appeared to be almost perfect. The gynoecium in general seems to be the most affected and the calyx the least. The average number of parts per floral whorl is seen to be progressive from the outer toward the innermost. There is a striking trend toward lower locule number as the season advances. This was exhibited in every one of the varieties. Analysis of variance indicated that the differences in mean locule number between months within a variety as were those between varieties, were highly significant. The end-of-season effect was most pronounced in the highly fasciated varieties; this was clearly brought out by a highly significant interaction between varieties and months. In comparing pollination and fruit set values Red Pear as a pollen parent gave the highest percent of sets and Santa Clara Canner, among the fasciated varieties, set the highest percent. What significance can be attached to this is uncertain. The F1 generation in every instance gave evidence that non-fasciation in tomatoes is partially dominant. The average F1 locule number always exceeded the average of the non-fasciated parent. The association between the F1 plants was almost proportional to the magnitude of their fasciated parental values. Red Currant (L. pimpilzellijolium) and Sugar (L. esculentum) show the highest degree of dominance for non-fasciation. Least dominance was exhibited by Red Peach and Yellow Peach, both possessing the highest locule number among the non-fasciated parents. Since these varieties are smooth, oval and regular in fruit contour one must presume that they possess either or both Lc and F genes. Yet they differ markedly in their degree of dominance when crossed with strictly fasciated varieties. A noticeable difference existed among the fasciated varieties with respect to the degree of modification of their F1 progeny. Marglobe and Pritchard were most affected. The highly fasciated varieties such as Brimmer, Colossal, and Baldwin I1 were modified only slightly. This indicates that in addition to one or two major recessive fasciation genes, they may possess numerous modifying genes that determine the degree to which the character is expressed.

ACKNOWLEDGMENTS The major portion of this work was carried on under direction of DR. ORLANDE. WHITE, to whom much credit is due for helpful criticism. Appreci- ation is expressed for the valuable assistance of DR. R. A. SILOW,Visiting Professor of Genetics, THE BLANDYEXPERIMENTAL FARM, UNIVERSITY OF VIRGINIA. 428 QUENTIN B. ZIELINSKI This investigation was supported in part by a research fellowship grant by the GENERALEDUCATION BOARD.

LITERATURE CITED DEVRIES,H., 1894 Over de Erfelijkheid der Fasciatien. Bot. Jaarboek 6: 72-118. FRIMMEL,FRANZ, 1922 Uber die Vererbung der Fruchtgrosse der Tomaten. Z. Pflanzenz. 8: 457- 462. GODRON,A., 1871-72 Melanges de TCratologie VCgBtale. MBm. SCC.nat. Sci. Cherbourg 16: 81- 127. HOUGHTALING,HELEN B., 1935 A developmental analysis of size and shape in tomato fruits. Bull. Torrey Bot. Club. 62: 243-252. Hns, H., 1906 Fasciation in Oxalis cvennta and experimental production of fasciation. Rpt. MO. Bot. Gard. 17: 147-152. KNOX,A. A., 1908 Induction, development and heritability of fasciations. Carneg. Inst. Pub. No. 98, 20 pp. LESLEY,J. W. and ROSA,J T., 1926 The improvement of tomatoes by selection. Hilgardia 2: 25-45. LINDSTROM,E. W., 1927 The inheritance of ovate and related shapes of tomato fruits. J. Agric. Res. 34: 961-985. 1928 Linkage of size, shape, and color genes in Lycopersicon. Z.I.A.V., Supplementb. 2: 1031-1057. MACARTHUR,J. W., 1926 Linkage studies with the tomato. Genetics 11 : 387-405. 1934 Linkage groups in the tomato. J. Genet. 29: 113-133. 1944 Personal communication. September 7. MASTERS,M. T., 1869 Vegetable Teratology. 534+xxxviii pp. London. (pp. 1-38). MOLLIARD,M., 1897 Hypertrophy pathologique des cellules vCgCtales. Rev. Gen. de Rot. 9: 33. MOQUIN-TANDON,A., 1841 Elements de TCratologie VCgCtale. Paris. (pp. 1-12). POWERS,LEROY, 1939 Formulas for determining theoretical effects of certain genetic factors upon inheritance of quantitative characters, with special reference to a study of a Lycopersi- con hybrid. J. Agric. Res. 59: 555-579. WARREN,PAUL A., 1924 Genetic studies in Lycopersicon. I. The heredity of fruit shape in the garden tomato. Mich. Acad. Sci., Arts and Letters paper 4: 357-394. WHITE,ORLAND E., 1913 The bearing of teratological development in Nicotiana on theories of heredity. Amer. Nat. 47: 205-228. 1916 Studies of teratological phenomena in their relation to evolution and the problems of heredity. ZIAV 16: 49-185. 1945 The biology of fasciation. J. Hered. 36: 11-22. 1948 Fasciation. Bot. Rev. (In press). YEAGER,A. F., 1937 Studies on the inheritance and development of fruit size and shape in the tomato. J. Agric. Res. 55: 141-152. ZIELINSKI,QUENTIN B., 1945 Fasciation in horticultural plants with special reference to the tomato. Amer. Soc. Hort. Sci. 46: 263-268.